Bo Di Muccio writes about how technology companies can turn AI into business results, with a focus on professional services, forward-deployed engineering, and pricing work around customer outcomes.
Bo Di Muccio is most useful for explaining how AI changes professional services models, forward-deployed engineering, measurement, and outcome-based pricing.
Best for
It fits a B2B SaaS company with a growing services arm taking AI-enabled offers to market.
Not for
It does not fit a developer-tools company launching a self-serve API and seeking demo videos and signups.
Key takeaways
AI widens the services layer because models do not deliver business outcomes on their own.
Forward-deployed engineering is a distinct operating model, not a renamed professional-services bucket.
Preliminary TSIA survey found outcome-based pricing obstacles were internal to providers, not customer reluctance.
Check first: Confirm cited figures and preliminary survey data, plus audience breakdown and speaking fee.
Main ideas
AI widens the services layer because models do not deliver business outcomes on their own · Forward-deployed engineering is a distinct operating model, not a renamed professional-services bucket · Outcome-based pricing is held back by the provider's measurement problem
Bo Di Muccio writes about how technology companies turn AI into business results, especially through professional services and forward-deployed engineering.
Their central argument: powerful models do not deliver value by themselves; people and teams must adapt, deploy, and measure the work.
They read company announcements for clues about how services are changing—from embedded engineering to repeatable AI solutions—and make the business implications concrete.
Di Muccio connects broad claims about AI to practical choices for services firms.
They also distinguish advisory work from forward-deployed engineering, arguing that the two carry different risks, margins, skills, and measures of success (their explanation of the two models).
A second recurring idea is that AI can reduce delivery effort without reducing customer value.
Their ServiceNow example shifts attention from hours worked to adoption and outcomes.
In a post describing preliminary TSIA survey results, they say most cited obstacles to outcome-based pricing were internal to providers—not customer reluctance—including difficulty measuring results, inconsistent delivery, finance constraints, and misaligned incentives (their pricing analysis).
Readers interested in AI services strategy can follow their work for analysis connecting industry moves to operating models, measurement, and pricing decisions.
Bo Di Muccio's core argument is that AI makes the services layer more important, not less, because a capable model does not by itself produce a business result.
Read more
The value sits in readiness, integration, adoption and embedded engineering.
They argue that services leaders misprice that work when they treat advisory projects, forward-deployed engineering and repeatable products as one business.
A second priority is pricing: outcome-based pricing stalls mainly because providers cannot measure and standardize outcomes, not because customers refuse the idea.
The posts are short, news-pegged arguments that use company announcements and TSIA survey results as evidence.
AI widens the services layer because models do not deliver business outcomes on their own
Bo Di Muccio takes the loudest prediction in tech — that AI collapses the services layer and pools value in the model — and points to where the money went instead.
Read more
Anthropic launched a services firm, OpenAI acquired consulting firms, and Microsoft, AWS and Google built similar capabilities, with more than $9 billion committed in about two months.
The reasoning: the distance between a capable model and an actual business result is enormous, and everything that closes that distance is services.
They cite the claim that roughly 95% of enterprise AI pilots produce no measurable P&L impact, then separate advisory work before and after the build from the embedded build work itself.
The OpenAI–IBM deal is presented as evidence: it was not a sale of API credits but a deal that put IBM consultants and forward-deployed engineers in the field, trained through the OpenAI Partner Network, deploying OpenAI models inside banks, telcos, retailers and government agencies. Bo Di Muccio asks why a model maker would rent distribution, integration and deployment muscle if the model were the whole game.
Forward-deployed engineering is a distinct operating model, not a renamed professional-services bucket
Bo Di Muccio draws a firm line between advisory professional services and forward-deployed engineering.
Read more
Advisory work covers readiness, data and operating-model fit, and governance after launch.
Forward-deployed engineers are embedded with the customer, work on the customer's data, own the outcome, and stay aligned to the product and go-to-market motion.
These carry different risks, margins, talent requirements and measures of success, they argue, so collapsing them into one bucket called PS or FDE leads services leaders to misprice both.
The posts cite OpenAI's Deployment Company launching with 150 forward-deployed engineers and $4 billion behind it, and the OpenAI–IBM partnership putting IBM consultants and forward-deployed engineers in the field. Both show that the build function is being funded as its own operating model, not as spare consulting capacity.
Outcome-based pricing is held back by the provider's measurement problem
Bo Di Muccio challenges the standard explanation that customers block outcome-based pricing because they are not ready or do not like the math.
Read more
In preliminary TSIA survey results described in the posts, only about one in five professional services leaders named the customer as the biggest barrier to outcome-contingent pricing.
Roughly four in five named internal issues: difficulty measuring outcomes, delivery that is not standardized enough to contract against, finance or revenue-recognition constraints, and misaligned incentives.
The most common answer was difficulty measuring and attributing outcomes to the provider's own work.
They conclude that the constraint sits largely on the provider's side of the table, making it actionable.
The survey asks for the single biggest barrier to outcome-contingent pricing.
The answers are revealing: most respondents describe their own measurement, standardization, finance or incentive problems rather than customer resistance.
The posts describe the results as preliminary, with responses still open.
The shift is from billable hours to adoption and outcomes
Bo Di Muccio uses ServiceNow's internal AI deployment to argue that AI compresses the hours required to deliver a service, not the value of the result.
Read more
The old software services organization made money mostly on implementation — bodies, weeks and a statement of work measured in hours.
Under what the posts call PS 2.0, services organizations make money on adoption and outcomes.
Deleting billable hours does not automatically mean lost revenue if the same work produces more value per customer over the life of a contract.
The posts say ServiceNow put AI agents on its own customer and employee operations and removed about 2.5 million work hours. Bill McDermott of ServiceNow, as described in the post, pointed instead to customers growing revenue per seat several times over across the life of a contract: fewer hours going in, much more value coming out.
Services firms should productize repeatable solutions and ship solutions, not people
Bo Di Muccio argues that the important detail in Accenture and Google Cloud's agentic AI unit is not the headline headcount of 1,000 forward-deployed engineers and 50,000 cloud-skilled people behind them.
Read more
The unit's charter, as described in the posts, is to build replicable, industry-specific solutions and stand up capability centers that turn experiments into transformation.
Productize an agent once and deploy it repeatedly across clients.
That shifts the commercial asset from effort expended or hours billed to a result that can be packaged, priced, scaled and resold.
They describe the result as a product company that happens to deliver services, and suggest watching whether earnings language moves from people delivered to solutions shipped.
The post cites a deployment that cut average handle time by more than a third.
The result matters because it can be packaged and reused, unlike a bespoke project whose value is tied to a particular team and client.
The Accenture unit exemplifies a large services firm re-chartering itself around repeatable products rather than bespoke statements of work.
This is the clearest statement of the argument running through the other posts.
Read more
Open it first to see exactly what Di Muccio claims and the evidence offered, before turning to pricing and vendor examples.
The post confronts tech’s loudest prediction — that AI collapses the services layer and pools value in the model — with evidence of where the money went.
Anthropic launched a services firm, OpenAI acquired two consulting firms, and Microsoft, AWS and Google built versions of the same thing, with more than $9 billion committed in about two months.
Di Muccio explains the gap between a capable model and a business outcome: by some measures, roughly 95% of enterprise AI pilots produce no measurable P&L impact, so model makers must fund the last mile themselves.
The sharpest move is a split.
Advisory professional services cover readiness, data and operating-model fit, and governance after launch.
Forward-deployed engineering is different work: engineers embedded with the customer, working on the customer's data, owning the outcome and aligned to product and go-to-market.
The post says these carry different risks, margins, talent requirements and measures of success.
Lump them into one bucket called PS or FDE, and services leaders misprice both.
For brands planning a campaign
Gives a keynote its spine: a B2B SaaS services organisation can argue that AI-era demand sits in the deployment layer, and that advisory and embedded build work need separate economics, not one blended rate card.
It answers a newcomer’s question — why hasn't outcome-based pricing scaled in twenty years? — with an actionable diagnosis that overturns the usual answer.
Read more
Di Muccio sets out the standard explanation for twenty years of slow adoption: buyers are not ready or dislike the maths when they see it.
Preliminary results from a TSIA survey of professional services leaders tell a different story.
Asked for the single biggest barrier to outcome-contingent pricing, only about one in five respondents pointed to the customer.
Roughly four in five described something inside the provider's own operation: difficulty measuring outcomes, delivery that is not standardised enough to contract against, internal finance or revenue-recognition constraints, and misaligned incentives.
The most common answer by a clear margin was difficulty measuring and attributing outcomes to the provider's own work.
Di Muccio's conclusion is useful: the constraint sits largely on the provider's side of the table, where the provider can act on it.
For brands planning a campaign
Use it as the diagnostic frame for an advisory session or workshop: before a SaaS services team renegotiates contracts around outcomes, it should audit measurement, delivery standardisation, and finance or rev-rec rules — the barriers its own leaders named.
It shows the method as much as the argument: take one announcement, read past the headline and see what it implies about who does the work.
Read more
Useful for anyone looking to apply the technique to their own market.
Di Muccio walks through OpenAI's deal with IBM and stresses what it was not: a sale of API credits.
The deal put IBM's consultants and forward-deployed engineers in the field, trained through the OpenAI Partner Network and deploying OpenAI models inside banks, telcos, retailers and government agencies.
The post also notes that OpenAI's own Deployment Company launched with 150 forward-deployed engineers and $4 billion behind it.
The reasoning is a simple inversion: if the model were the whole game, the model vendor would just sell it.
Yet the company with one of the most widely adopted models on earth concluded it could not drive customer outcomes alone and rented distribution, integration and deployment muscle from the likes of IBM and Accenture.
Hence the question posed to readers: why would AI companies race to build services capabilities if AI were about to gut services?
For brands planning a campaign
Supplies a concrete, checkable talking point for a keynote: model vendors are funding deployment capacity, making integration and adoption the defensible parts of a services offer, rather than access to a model.
The clearest worked example of the shift from hours to outcomes, with a named company, specific numbers and a direct answer to whether efficiency destroys a services business.
Read more
Di Muccio examines Bill McDermott's remarks at an investor conference, where the ServiceNow chief described putting AI agents on the company's own customer and employee operations and removing something like 2.5 million work hours.
The obvious reading is that a software services organisation deliberately deleted its own billable hours.
Di Muccio argues those hours did not register as lost revenue: McDermott pointed instead to customers growing revenue per seat several times over across the life of a contract, so fewer hours go in while much more value comes out.
The old software services organisation made money mostly on implementation — bodies, weeks and a statement of work measured in hours.
Under what the post calls PS 2.0, services organisations make money on adoption and outcomes.
The compression hits the effort, not the value of the result.
For brands planning a campaign
Gives a services leader the narrative for an internal or stage argument: if delivery hours fall, an organisation that still prices implementation hours absorbs the whole impact, while one that prices adoption and outcomes captures the value it creates.
This is the sell-side version of the same shift — a large services firm re-chartering itself — and it ends with a prediction readers can check against public reporting.
Read more
On Accenture and Google Cloud standing up an agentic AI unit, Di Muccio argues that the headline numbers — 1,000 forward-deployed engineers, 50,000 cloud-skilled people behind them — are the wrong story.
What matters is the charter: build replicable, industry-specific solutions and stand up capability centres that turn experiments into transformation, so a productised agent is built once and deployed repeatedly across clients.
The post cites a deployment that cut average handle time by more than a third, and explains the commercial significance: the result can be packaged, priced, scaled and resold, unlike effort expended or hours billed.
The conclusion: a firm whose best asset is a solution shipped a hundred times is a product company that happens to deliver services.
Di Muccio flags the imminent full-year results and a language test to watch: people delivered versus solutions shipped.
For brands planning a campaign
Gives a SaaS services organisation a benchmark question for its roadmap — how many offers are reusable products with a price, and how many are bespoke statements of work — plus a way to read peer earnings calls for signs of the same shift.
Argument-led posts built on market news and TSIA research5
Tight evidence base with implementation gaps5
Forward-deployed engineering as a separate operating model2
AI is widening the services layer rather than erasing it2
Outcome-based pricing begins with the provider's own measurement problem1
PS 2.0: price adoption and outcomes, not hours1
Productized services and repeatable solutions1
Forward-deployed engineering as a separate operating model
Bo Di Muccio draws a firm line between forward-deployed engineering and traditional professional services.
Read more
Forward-deployed engineers are embedded with the customer, work on the customer's data, own the outcome, and stay aligned to the product and go-to-market motion.
They argue that collapsing the two into one bucket misprices both because they carry different risk, margin, talent requirements, and measures of success.
The posts cite OpenAI's Deployment Company launching with 150 forward-deployed engineers and an OpenAI partnership that put IBM consultants and forward-deployed engineers in the field.
For brands planning a campaign
A B2B SaaS company can use this distinction to decide whether to build a separate forward-deployed engineering function, with its own hiring, delivery, and pricing model, rather than fold it into professional services. It gives a keynote or advisory session a concrete organizing question: which work is advisory, which is embedded delivery, and how should each be measured?
AI is widening the services layer rather than erasing it
Bo Di Muccio's recurring counter-thesis is that capable models do not automatically produce business outcomes.
Read more
The posts point to Anthropic launching a services firm, OpenAI acquiring consulting firms, and Microsoft, AWS, and Google building similar capabilities, with more than $9 billion committed in about two months.
They also cite the claim that roughly 95% of enterprise AI pilots produce no measurable P&L impact, then argue that the distance between a model and a business result is where services work lives.
The OpenAI-IBM partnership is presented as further evidence that a model maker will rent deployment and integration capacity rather than sell a model alone.
For brands planning a campaign
This is a useful frame for a keynote aimed at technology services leaders: AI creates demand for help putting models to work. Treat the dollar and pilot-impact figures as claims from the posts, and verify them against current primary sources before repeating them in campaign material.
Outcome-based pricing begins with the provider's own measurement problem
Bo Di Muccio challenges the common explanation that customers block outcome-based pricing.
Read more
In preliminary TSIA survey results described in the posts, only about one in five professional services leaders named the customer as the biggest barrier to outcome-contingent pricing.
Roughly four in five named internal issues: difficulty measuring and attributing outcomes, delivery that is not standardized enough to contract against, internal finance or revenue-recognition constraints, and misaligned incentives.
The most common answer was difficulty measuring and attributing outcomes to the provider's own work.
For brands planning a campaign
A workshop or advisory session can start with a provider-side readiness audit before discussing pricing structures.
That is more useful to a B2B SaaS services team than a generic outcome-pricing pitch.
The posts describe preliminary survey results, so disclose that status and do not present the percentages as a completed study.
Bo Di Muccio uses ServiceNow's internal AI deployment to illustrate the shift from implementation work to outcome work.
Read more
The posts say ServiceNow put AI agents on its own customer and employee operations and removed about 2.5 million work hours.
Rather than treating that as lost services revenue, they highlight Bill McDermott's emphasis on customers growing revenue per seat over the life of a contract.
The argument is that the old software services organization sold bodies, weeks, and statements of work measured in hours; the PS 2.0 organization sells adoption and outcomes.
AI compresses the hours required but not the value of the result.
For brands planning a campaign
This gives a B2B SaaS company a clear keynote angle: redefine the services scorecard around adoption and customer outcomes instead of billable hours. Present the ServiceNow example as an illustration, not proof that every services organization can remove hours without revenue loss.
Bo Di Muccio argues that the important detail in the Accenture and Google Cloud agentic AI unit is not the headline headcount of 1,000 forward-deployed engineers and 50,000 cloud-skilled people.
Read more
The unit's charter, as described in the posts, is to build replicable, industry-specific solutions and capability centers that turn experiments into transformation.
Productize an agent once and deploy it repeatedly across clients.
They call this a move from bespoke projects to repeatable products, or from delivering people to shipping solutions, and describe the result as a product company that happens to deliver services.
For brands planning a campaign
For a SaaS services strategy session, this supports a practical discussion about packaging repeatable offers, setting reuse targets, and measuring solutions shipped rather than headcount deployed. It also gives a campaign a concrete way to talk about operating-model change without relying only on abstract AI claims.
Argument-led posts built on market news and TSIA research
The posts follow a recognizable pattern: open with a current vendor announcement or an industry claim, state the counterintuitive interpretation, then support it with a TSIA survey result, an operating-model distinction, or both.
Read more
Recurring lines include "AI is the era of services" and "TSIA told you so."
The writing assumes a reader who knows professional services, forward-deployed engineering, statements of work, and revenue recognition.
The format is short and argument-led rather than a full research report or implementation guide.
Bo Di Muccio analyzes public company announcements and survey results, then reframes them into an operating-model argument for services leaders.
For brands planning a campaign
A brand can evaluate Bo Di Muccio for executive-level keynote or advisory content that responds quickly to market news and gives services leaders a strong point of view. Specify the audience's familiarity with services economics and ask for supporting data beyond the short posts.
The five posts repeat a focused set of claims about AI, professional services, forward-deployed engineering, and pricing.
Read more
Before booking, ask for the full survey methodology and sample composition, publication dates, and a completed client case; the posts do not provide these, or a full outcome-pricing model, staffing plan, or step-by-step operating-model transition.
The five short LinkedIn posts here do not establish the full range of Di Muccio's research, speaking topics, or client work.
For brands planning a campaign
Pair Bo Di Muccio's framing with the brand's own data, delivery examples, or a structured workshop to supply implementation detail. Do not promise a plug-and-play pricing model or a complete transformation playbook from the posts alone.
Bo Di Muccio's posts identify TSIA as the organization where they lead research and publish analysis on professional services and forward-deployed engineering.
Read more
The other companies named in the posts here — Anthropic, OpenAI, IBM, Microsoft, Accenture, ServiceNow, Google Cloud and AWS — appear as subjects of commentary and market analysis, not as stated sponsors, clients, or partners. The posts contain no sponsorship, affiliate, investment, or client disclosure; ask whether any of these relationships exist.
TSIA
Employment
Di Muccio writes as a TSIA researcher and analyst.
Read more
Their posts describe leading TSIA's research on forward-deployed engineering as a distinct operating model, cite a TSIA survey of professional services leaders on outcome-based pricing, and close with the recurring line "TSIA told you so" alongside the #TSIA hashtag.
Bo Di Muccio’s posts form a connected argument about AI and technology professional services.
Read more
Start with the pieces distinguishing forward-deployed engineering from advisory professional services, then read the outcome-based pricing post and the Accenture and ServiceNow examples.
The recurring question is where value sits once a model is capable: in readiness, integration, adoption and embedded delivery, not in the model alone.
The preliminary TSIA survey post offers pricing evidence.
Both write short LinkedIn arguments about how AI is changing technology professional services.
Read more
Thomas Lah argues that AI reprioritizes services rather than eliminating them, describes services as the engine of AI adoption, and discusses forward-deployed engineers and professional services economics. Bo Di Muccio covers the same territory.
Thomas Lah often builds a post around a linked third-party article or podcast episode, extending the discussion to revenue forecasting and sales-finance operations. Bo Di Muccio focuses on operating-model distinctions between advisory services, forward-deployed engineering and productized services, with a TSIA survey supporting the pricing discussion.
Both address how professional-services firms grow through client relationships and repeatable methods.
Read more
Rory Channer’s posts describe the Activator research and book: a framework for rainmaker behaviors, client engagement and firm-wide development, based on work with more than 100 professional services firms. Bo Di Muccio also writes for services leaders seeking to change how they sell and deliver.
Rory Channer focuses on business development and client-relationship behavior in law, accounting, consulting and similar firms.
Bo Di Muccio focuses on AI-era delivery and pricing models in technology services.
Channer’s posts announce books and research; Bo Di Muccio’s make news-pegged arguments.
Both describe an embedded, outcome-accountable operator rather than a scoped project team.
Read more
Eric Quanstrom explains a GTM Engineer role at Apollo that is accountable for customer pipeline growth, works without a statement of work or hourly clock, and follows a seven-pillar framework. Bo Di Muccio’s argument for forward-deployed engineering similarly distinguishes embedded delivery from traditional professional services.
Eric Quanstrom focuses on go-to-market engineering, pipeline generation, account cadence and multi-surface software distribution. Bo Di Muccio writes at the industry level about AI services economics, advisory versus build work, and outcome-based pricing.
For a B2B SaaS campaign aimed at professional-services leaders focused on AI-era go-to-market, Bo Di Muccio is a specialist placement, not a broad SaaS-audience buy.
Read more
Their posts, dated 23 June to 29 September 2026, address professional-services, customer-success, and field-deployment leaders at technology vendors, using terms such as “PS 2.0,” forward-deployed engineering, and outcome-based pricing.
The posts show 2,097 followers but provide no breakdown by job function, seniority, geography, or engagement.
Before booking, ask for an audience survey or documented breakdown covering job function, seniority, and company type, plus recent results for comparable services-strategy posts.
LinkedIn followers
2,097
LinkedIn profile follower count at observation · 2026-10-11
The five posts contain no sponsorship offering, rate card, media kit, or partnership inquiry route, so there are no published formats to list.
Read more
Bo Di Muccio writes in TSIA’s research voice, arguing that AI widens rather than shrinks the professional services layer, forward-deployed engineering is a distinct operating model from advisory professional services, and providers should price and operate around customer outcomes instead of hours billed.
The posts reference a TSIA survey of professional services leaders and a year spent leading TSIA’s research on forward-deployed engineering.
A partnership would need to be scoped from scratch.
The posts don’t specify deliverables, disclosure practice, audience size for sponsored placements, production timelines, or commercial terms.
Treat this as an editorial and research relationship, not a packaged creator buy.
The nearest documented commercial home for this work is TSIA, whose research platform and events the posts repeatedly point to.
A team seeking to reach professional services and customer-success leaders through this perspective should ask TSIA directly about sponsorship, research-partnership, or event options before including them in a plan; the posts don’t confirm that route.
If a personal partnership offer exists, ask Bo or TSIA before quoting it.
A B2B SaaS company with a growing services arm is taking AI-enabled offers to market and needs to rebuild its professional-services go-to-market around them. It wants a keynote and an advisory working session for about 40 services, customer-success and delivery leaders on AI-era services models and outcome-based pricing, with two live offers re-scoped by the end.
Read more
Reasons
The brief's central question is the one these posts argue about: how should a SaaS services team selling AI work distinguish advisory services—readiness, data and operating-model fit, and governance after launch—from embedded delivery—engineers working on the customer's data, owning the outcome, and aligning with product and go-to-market? The posts argue that the two models carry different risks, margins, talent needs and success measures, and that blending them misprices both. This gives the keynote an organising idea beyond a tour of AI trends and gives the room a decision to make about its org chart and rate card.
The pricing material turns the advisory session into a diagnostic. The published read of preliminary TSIA survey results is that only about one in five professional services leaders named the customer as the biggest barrier to outcome-contingent pricing, while roughly four in five pointed to provider-side barriers: difficulty measuring and attributing outcomes, delivery too unstandardised to contract against, internal finance or revenue-recognition constraints, and misaligned incentives. These four barriers make a ready-made working-session agenda; measurement, the most common answer, is something the room can start addressing that day.
Named, checkable cases can anchor the argument: ServiceNow removing about 2.5 million work hours from its own operations while pointing to revenue per seat over a contract's life; Accenture and Google Cloud chartering an agentic AI unit to build replicable industry solutions and capability centres rather than bespoke projects, with a cited deployment that cut average handle time by more than a third; and OpenAI partnering with IBM to put consultants and forward-deployed engineers in the field. Each gives services leaders a story to map onto their own business, grounding the argument that AI widens the services layer.
A speaker with a stated thesis is easier to programme than a neutral analyst. “AI is the era of services” and the hours-to-outcomes argument give the event a point of view. The recurring technique of reading an announcement for what it implies about who does the work is one the audience can use to assess its own market.
Conditions
The figures in the posts are claims, not independently verified here. Before using them on a slide, check the roughly 95% pilot-impact number, the more than $9 billion committed in about two months, the 2.5 million ServiceNow hours, the Accenture headcounts and handle-time reduction, and the OpenAI Deployment Company's 150 engineers and $4 billion against current primary sources.
The outcome-pricing percentages are preliminary survey results, with responses still open. Present them as preliminary and include the sample and fieldwork dates if TSIA can supply them; otherwise, drop the percentages and use the four provider-side barriers as the discussion frame.
Bo Di Muccio's posts offer no implementation playbook: no full outcome-pricing model, staffing plan, step-by-step transition or completed client case. Pair the speaker with the brand's delivery cost data, contract templates and revenue-recognition rules, and have a facilitator ensure the session produces artefacts, not just notes.
Before booking, ask Bo Di Muccio or TSIA for the speaking fee, deliverables, disclosure practice and inquiry route; none is documented.
Treat this as a specialist placement, not a reach buy. Bo Di Muccio has 2,097 followers, but no breakdown by job function, seniority, company type or engagement. Paid distribution to a large developer or buyer audience is unsupported. The value is the argument and the room, not impressions.
Verdict
Strong fit on subject and argument, provided the brand supplies its own data and verifies the figures before they go on a slide.
When to choose a specialist
A developer-tools company launching a self-serve API wants signups from working software engineers. It wants three short demo videos and a promotional code, distributed as paid creator posts on LinkedIn and YouTube over two weeks, with the creator's fee tied to registrations.
Read more
Reasons
Bo Di Muccio's five posts are written, argument-led commentary based on vendor announcements and survey results; they show no video, demo, tutorial or walkthrough work. The posts do not support a three-video demo package, and the two-week turnaround would need to be agreed from scratch.
The writing targets services, customer-success and field-deployment leaders at technology vendors. It assumes familiarity with statements of work, revenue recognition, delivery margin and forward-deployed engineering as an operating model. Rank-and-file engineers deciding whether to try an API are a different audience with a different question, and the posts offer no evidence of reaching them.
Before considering a performance-tied, two-week paid placement, ask for a rate card, media kit, deliverables, disclosure practice, audience breakdown and inquiry route. None is documented in Bo Di Muccio's posts, which also contain no partnership invitation, so the offer would need to be built from scratch rather than bought as a fast launch placement.
The message conflicts with the brief. The core argument is that a capable model or product is close to worthless until someone deploys it, and that value sits in the services layer around the tool. Selling a self-serve API on the promise that engineers can skip that layer would put the creator at odds with their published position, a poor fit for the audience that does follow them.
Registration targets cannot be underwritten from the available data: there is only a follower count, with no performance history for comparable posts. Ask for that history before setting targets.
Verdict
Reject for this brief.
A worked campaign example
Angle
Run the engagement as a working session for a SaaS services leadership team, not a stage talk with questions at the end. By the close, two live offers should have a defined outcome unit, measurement method and contract trigger, and the team should be able to distinguish advisory work from embedded delivery.
Campaign title
Reprice the last mile: a keynote and offer-repricing session
Distribution note
Treat any publishing, recording or paid distribution around the session as a separate agreement. No sponsorship format, rate or disclosure practice is documented; agree rights to record, repurpose or promote the material separately.
How to judge whether it worked
Every participant can sort the team's offers into advisory versus embedded delivery and explain why.
Two one-page offer sheets are complete, each with a named outcome unit, measurement method and contract trigger reviewed by finance.
The team can identify which of the four provider-side barriers applies to them and name the first fix.
At 30 days, internal reporting and pipeline reviews use outcome language for at least one offer, and the brand can point to a specific change in how it prices or scopes work.
Run of show
Before the session, the brand sends two live AI-related offers with their current pricing model, one delivery-cost dataset, one contract template and its revenue-recognition constraints, plus the metric it wants to move (for example, share of revenue priced on outcomes rather than hours).
Opening, 30 minutes: the bet against the services layer and where the money went instead. The split between advisory professional services and forward-deployed engineering, and why blending them misprices both. Apply the distinction to the brand's two offers: which bucket does each belong in, and does current pricing reflect that?
Second block, 30 minutes: the four provider-side barriers to outcome-contingent pricing. Poll the room to focus on the barriers the team actually faces, rather than the ones the deck assumes.
Workshop, 60 minutes: tackle one offer at a time. Define the outcome in the customer's terms, how it will be measured and attributed, what must be standardised in delivery before it can be contracted, and whether the trigger meets finance and revenue-recognition rules, with the finance decision-maker in the room. The output is a one-page offer sheet for each offer.
Close and handoff, 15 minutes: agree on a 30-day check of language and metrics, and a test for public evidence of the shift suggested by the posts: whether internal reporting starts describing solutions shipped rather than people delivered.
What the brand must supply
Delivery-cost and margin data for both offers, so repricing is grounded in the brand's numbers rather than the speaker's examples.
A finance or revenue-recognition decision-maker with authority to say no in the room.
An executive sponsor to own the 30-day follow-up.
Verified versions of any external figures used in the deck, or agreement to use only the brand's data and unnamed patterns.
Choose Thomas Lah when the services arm sits inside a wider revenue organisation and the sponsor wants the keynote to connect services economics to forecasting and margin strategy, not only to service delivery.
Read more
The posts cover services as the engine of AI adoption and the argument for trading early services margin for lock-in and lifetime value, then move to revenue forecasting going model-driven rather than meeting-driven, with sales conversations, account data and operating data pulled into one live view.
That is a useful second half for a company whose services leaders share a room with sales and finance.
Lah is also the better choice if the format is a recorded interview or podcast rather than a stage talk, since the forecasting argument is presented through a guest episode and paired with a finance-side source.
The AI-services argument draws mainly on other people's work: a venture firm's article on services-led growth and an AI company finance chief's post on forecasting.
It's easy to follow but offers little diagnostic detail on why outcome-based pricing stalls, provider-side barriers, or how to name an outcome unit, measure attribution and set a contract trigger.
A team repricing two live offers would need to ask for the four-barrier frame and repricing worksheet.
The margin-for-moat point stays at the level of principle, so a brand would need to build its own numbers.
Choose Rory Channer when the constraint is partners' ability to win and hold client work, rather than offer pricing.
Read more
The posts draw on a book published by Harvard Business Review Press in May 2025, based on more than 18 months of research and input from over 100 professional services firms.
They describe named client-engagement behaviours a firm can train, rather than a thesis to agree with.
That suits a partnership-led audience in law, accounting or consulting, or a services organisation whose delivery is fine but whose client relationships are static.
It also suits a longer engagement: the posts point to a development programme and partner retreats, with pre-orders and bulk orders offered, making a multi-session rollout a closer fit than a single keynote.
The posts focus on how rainmakers behave with clients and how firms build that capability; they do not address AI-era delivery models, forward-deployed engineering, outcome-based pricing or outcome measurement.
For a campaign that needs two live offers with a named outcome unit, a measurement method and a contract trigger reviewed by finance, Channer can help change how the room sells, but not how it prices or scopes.
The engagement is also different: this reads as a firm-wide capability programme sold to a partnership, not a one-off keynote plus working session for 40 leaders.
A brand seeking a single event should confirm that format is available.
Choose Eric Quanstrom when the brand wants to emulate an embedded, outcome-accountable operator rather than an advisory engagement.
Read more
The posts describe the role from inside: accountability for whether the customer produces more pipeline three months out; embedding with the customer's team for the contract term, with no statement of work and no clock on hours; and a fixed seven-step sequence from total addressable market through to reporting.
The work spans roughly 100 accounts, with eight to ten meetings per account per year, and pay tied to account health.
The posts also cover what happens when an AI-native product is used outside its own application, as users reach the tool through chat assistants, the CRM and the browser instead of logging in.
That makes Quanstrom useful if the brand's AI offers will be delivered by an embedded team or distributed across surfaces rather than sold as a destination product.
This is one vendor's go-to-market engineering practice in sales tooling, not cross-industry services economics.
It offers no pricing model, revenue-recognition or contracting discussion, survey of professional services leaders, or comparison of advisory work with build work.
The multi-surface figures are the company's own performance claims, not independent evidence.
The posts give a clear picture of what the embedded role does week to week and why the sequence matters, but nothing to put straight into an offer sheet.
For a brief ending with two repriced offers, Quanstrom is a practitioner's reference, not the framework; ask for independent validation of the performance claims.
Bo Di Muccio's five LinkedIn posts discuss AI and technology professional services, forward-deployed engineering, productized services and outcome-based pricing.
The posts are short, news-pegged arguments.
No publication dates appear in the posts, so they present recurring positions rather than a dated sequence.
The posts identify TSIA as the organization where Bo Di Muccio leads research and publishes analysis.
Company names such as OpenAI, IBM, ServiceNow, Accenture and Google Cloud appear as subjects of commentary, not as stated clients or sponsors.
The posts here include no full reports, talk transcripts, video, client cases, rate card or audience breakdown.
Figures and survey percentages appear as reported in the posts and were not independently checked.
The comparison with similar creators covers LinkedIn posts by Thomas Lah, Rory Channer and Eric Quanstrom.
